CIRCL: Daniel Mann (Phonology)
OCT 24, 2017 | 6:30 PM TO 8:00 PM
The Graduate Center
365 Fifth Avenue
October 24, 2017: 6:30 PM-8:00 PM
Daniel Mann presents on a topic in Phonology.
Title:Possible homologous structures in human and budgerigar syllables
Despite their seeming open-endedness, learned systems, like birdsong and human language, are subject to biases and constraints. The specific mechanisms, limits, and magnitude of these biases are still not well understood, though. In human language, limits on learning have often been attributed to innate knowledge that is specific to humans and to language. However, comparisons with non-human communication systems have provided ample evidence that many biases can be explained by more general mechanisms that are shared with non-human species.
The budgerigar has a communication system that provides an interesting comparison for humans. The small Australian parrot has a learned complex song that is highly variable, has very few stereotyped patterns, and has large amounts of modulation in several acoustic parameters. My colleagues and I used approaches and techniques from human speech analysis to create an algorithm which would divide budgerigar song. This algorithm divides the acoustic stream into units that are more analogous to human phones and syllables than the units typically used in comparative bio-acoustics.
Analysis of these units from three independent groups of budgerigars reveal that they share several sound pattern tendencies with humans. These patterns are most consistent with bio-mechanical explanations, particularly those related to phonation. Furthermore, these patterns support phonological theories that contend that (near) universal patterns are mostly emergent from articulatory, perceptual, or communicative pressures that are not specific to language or humans (such as Blevins’s Evolutionary Phonology).